Dear Patricia, Take a look at this post: http://www.nabble.com/Re:-applying-cor.test-to-a-(m,-n)-matrix---SUMMARY-to17150239.html#a17150239
The easiest way is using cor. See ?cor for details. Here is an example: set.seed(12) A<-matrix(rnorm(30),ncol=3) cor(A) # [,1] [,2] [,3] # [1,] 1.0000000 -0.2015320 0.7235916 # [2,] -0.2015320 1.0000000 -0.5559593 # [3,] 0.7235916 -0.5559593 1.0000000 HTH, Jorge On Mon, Feb 9, 2009 at 11:44 AM, patricia garcía gonzález < kurtney...@hotmail.com> wrote: > > Hi all, > > Having a matrix A formed by n vectors as columns. Is there anything to > calculate a determined function to all combination of vectors? > > For example imagine A matrix is compose by vectors a, b and c. And the > function to perform is correlation, so I would like to obtain cor(a, b), > cor(a, c) and cor(b, c). > > I we had numbers instead of vector, the function is outer, but I am not > able to apply it to vectors... > > Thanks a lot. > > P. > > > > > Date: Mon, 9 Feb 2009 08:37:06 -0800 > > From: zzn...@gmail.com > > To: r-help@r-project.org > > Subject: Re: [R] installing R on Ubuntu > > > > On Mon, Feb 9, 2009 at 4:51 AM, Neil Shephard <nsheph...@gmail.com> > wrote: > > > > > > The preceived "difficulty" of installing R under whatever flavour of > > > GNU/Linux in this thread stems from being unfamiliar with the process > of the > > > package management of the flavour of GNU/Linux you use (and in part by > the > > > various distros not having the most recent version of R in their > > > repositories. > > > > > > People who say "why can't it be as easy as dowloading a self-installing > > > binary and running that" are trying to fit a round peg (their > experience and > > > understanding of how applications install in M$-windows) in a square > hole > > > (or triangular, hexagonal, or whatever depending on the distribution of > > > GNU/Linux). > > > > This is true. However, for the most common Linux distros --Debian, Red > > Hat Enterprise / CentOS / Scientific Linux / Fedora, openSUSE and > > Ubuntu -- you can install the most recent R compiled for your distro > > from > > > > http://<your-nearest-CRAN-mirror>/bin/linux/ > > > > In addition, most of the distros have third-party repositories where > > you can find the latest version of R. In short, if you have an x86 or > > x86_64/amd64 system running almost any Linux, you can find a > > pre-compiled R. R is a popular package, and it's pretty easy to find > > even for Power PC or some of the obscure architectures. > > > > > > > > There are pro's and con's to each of the GNU/Linux flavours and its > really a > > > matter of deciding which you like/have invested time in learning. > > > > > > Irrespective its still simple to install R from source under > GNU/Linux... > > > > > > 1) Download source tar-ball > > > 2) Extract and cd to the directory > > > 3) ./configure --prefix=/where/you/want/R/to/go (optionally setting the > > > install path at this stage) > > > 4) ./make > > > 5) ./make install > > > > > > ...all documented in the FAQ at > > > > http://cran.r-project.org/doc/FAQ/R-FAQ.html#How-can-R-be-installed-_0028Unix_0029 > > > > Many Linux distros do *not* install the development tools by default, > > and which ones live in which packages varies by distro. Fedora in > > particular is extremely stripped when you install from the LiveCD. You > > have to install gcc, make and a couple of other things just to install > > VMware Tools, for example, when running Fedora as a VMware guest. For > > building R from source and installing R packages, you'll also need to > > install gfortran. And many libraries with external dependencies, like > > Rgraphviz, will require not only the package itself (graphviz) but > > also the C headers, which may have the name "graphviz-devel" on some > > distros and some other name on other distros. > > > > > > This might not be as clean as using the native package management, but > does > > > mean that you'll have the latest version installed. > > > > > > Neil > > > > > > (Addendum - I've tried several different distros, starting with RedHat > 7.3, > > > then various versions of Slackware 8 through to 9 before settling on > Gentoo, > > > all were easy to install R in). > > > > I just recently switched from Gentoo to openSUSE. Gentoo usually had > > the latest R source in their repository within a day or so of it > > coming out of the R Project release cycle. To get it, all you needed > > to do was put the package name in the "/etc/portage/package-keywords" > > file. And Gentoo, since it is almost all compiled from source, by > > nature *does* have all the development tools installed and installs > > all the headers when it installs packages. > > > > -- > > M. Edward (Ed) Borasky > > > > I've never met a happy clam. In fact, most of them were pretty steamed. > > > > ______________________________________________ > > R-help@r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > > and provide commented, minimal, self-contained, reproducible code. > > _________________________________________________________________ > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]]
______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
